2022-09-22 03:13:36

by Kees Cook

[permalink] [raw]
Subject: [PATCH 02/12] skbuff: Proactively round up to kmalloc bucket size

Instead of discovering the kmalloc bucket size _after_ allocation, round
up proactively so the allocation is explicitly made for the full size,
allowing the compiler to correctly reason about the resulting size of
the buffer through the existing __alloc_size() hint.

This will allow for kernels built with CONFIG_UBSAN_BOUNDS or the
coming dynamic bounds checking under CONFIG_FORTIFY_SOURCE to gain
back the __alloc_size() hints that were temporarily reverted in commit
93dd04ab0b2b ("slab: remove __alloc_size attribute from __kmalloc_track_caller")

Cc: "David S. Miller" <[email protected]>
Cc: Eric Dumazet <[email protected]>
Cc: Jakub Kicinski <[email protected]>
Cc: Paolo Abeni <[email protected]>
Cc: [email protected]
Cc: Greg Kroah-Hartman <[email protected]>
Cc: Nick Desaulniers <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Signed-off-by: Kees Cook <[email protected]>
---
net/core/skbuff.c | 34 ++++++++++++++++++----------------
1 file changed, 18 insertions(+), 16 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 974bbbbe7138..4fe4c7544c1d 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -427,14 +427,15 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
*/
size = SKB_DATA_ALIGN(size);
size += SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
- data = kmalloc_reserve(size, gfp_mask, node, &pfmemalloc);
- if (unlikely(!data))
- goto nodata;
- /* kmalloc(size) might give us more room than requested.
+ /* kmalloc(size) might give us more room than requested, so
+ * allocate the true bucket size up front.
* Put skb_shared_info exactly at the end of allocated zone,
* to allow max possible filling before reallocation.
*/
- osize = ksize(data);
+ osize = kmalloc_size_roundup(size);
+ data = kmalloc_reserve(osize, gfp_mask, node, &pfmemalloc);
+ if (unlikely(!data))
+ goto nodata;
size = SKB_WITH_OVERHEAD(osize);
prefetchw(data + size);

@@ -1709,6 +1710,7 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail,
{
int i, osize = skb_end_offset(skb);
int size = osize + nhead + ntail;
+ int alloc_size;
long off;
u8 *data;

@@ -1722,11 +1724,11 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail,

if (skb_pfmemalloc(skb))
gfp_mask |= __GFP_MEMALLOC;
- data = kmalloc_reserve(size + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
- gfp_mask, NUMA_NO_NODE, NULL);
+ alloc_size = kmalloc_size_roundup(size + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)));
+ data = kmalloc_reserve(alloc_size, gfp_mask, NUMA_NO_NODE, NULL);
if (!data)
goto nodata;
- size = SKB_WITH_OVERHEAD(ksize(data));
+ size = SKB_WITH_OVERHEAD(alloc_size);

/* Copy only real data... and, alas, header. This should be
* optimized for the cases when header is void.
@@ -6063,19 +6065,19 @@ static int pskb_carve_inside_header(struct sk_buff *skb, const u32 off,
int i;
int size = skb_end_offset(skb);
int new_hlen = headlen - off;
+ int alloc_size;
u8 *data;

size = SKB_DATA_ALIGN(size);

if (skb_pfmemalloc(skb))
gfp_mask |= __GFP_MEMALLOC;
- data = kmalloc_reserve(size +
- SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
- gfp_mask, NUMA_NO_NODE, NULL);
+ alloc_size = kmalloc_size_roundup(size + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)));
+ data = kmalloc_reserve(alloc_size, gfp_mask, NUMA_NO_NODE, NULL);
if (!data)
return -ENOMEM;

- size = SKB_WITH_OVERHEAD(ksize(data));
+ size = SKB_WITH_OVERHEAD(alloc_size);

/* Copy real data, and all frags */
skb_copy_from_linear_data_offset(skb, off, data, new_hlen);
@@ -6184,18 +6186,18 @@ static int pskb_carve_inside_nonlinear(struct sk_buff *skb, const u32 off,
u8 *data;
const int nfrags = skb_shinfo(skb)->nr_frags;
struct skb_shared_info *shinfo;
+ int alloc_size;

size = SKB_DATA_ALIGN(size);

if (skb_pfmemalloc(skb))
gfp_mask |= __GFP_MEMALLOC;
- data = kmalloc_reserve(size +
- SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
- gfp_mask, NUMA_NO_NODE, NULL);
+ alloc_size = kmalloc_size_roundup(size + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)));
+ data = kmalloc_reserve(alloc_size, gfp_mask, NUMA_NO_NODE, NULL);
if (!data)
return -ENOMEM;

- size = SKB_WITH_OVERHEAD(ksize(data));
+ size = SKB_WITH_OVERHEAD(alloc_size);

memcpy((struct skb_shared_info *)(data + size),
skb_shinfo(skb), offsetof(struct skb_shared_info, frags[0]));
--
2.34.1


2022-09-22 19:41:43

by Jakub Kicinski

[permalink] [raw]
Subject: Re: [PATCH 02/12] skbuff: Proactively round up to kmalloc bucket size

On Wed, 21 Sep 2022 20:10:03 -0700 Kees Cook wrote:
> diff --git a/net/core/skbuff.c b/net/core/skbuff.c
> index 974bbbbe7138..4fe4c7544c1d 100644
> --- a/net/core/skbuff.c
> +++ b/net/core/skbuff.c
> @@ -427,14 +427,15 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
> */
> size = SKB_DATA_ALIGN(size);
> size += SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
> - data = kmalloc_reserve(size, gfp_mask, node, &pfmemalloc);
> - if (unlikely(!data))
> - goto nodata;
> - /* kmalloc(size) might give us more room than requested.
> + /* kmalloc(size) might give us more room than requested, so
> + * allocate the true bucket size up front.
> * Put skb_shared_info exactly at the end of allocated zone,
> * to allow max possible filling before reallocation.
> */
> - osize = ksize(data);
> + osize = kmalloc_size_roundup(size);
> + data = kmalloc_reserve(osize, gfp_mask, node, &pfmemalloc);
> + if (unlikely(!data))
> + goto nodata;
> size = SKB_WITH_OVERHEAD(osize);
> prefetchw(data + size);

I'd rename osize here to alloc_size for consistency but one could
argue either way :)

Acked-by: Jakub Kicinski <[email protected]>